On Wed, Aug 10, 2016 at 5:42 PM, Goran Brkuljan <[email protected]> wrote: >> @Goran, did you let lxd create a new zfs pool on top of a file >> (loopback)? If yes (e.g. you're following >> https://help.ubuntu.com/lts/serverguide/lxd.html#lxd-configuration or >> similar) > > > Yes that's exactly what I did. > Thank you very much for the answer it works! > Anyway, production will probably wait one year. I am developing > prototype of some cloud platform, and containers are central point of > application. Still I don't know much about ZFS, I took it for granted > and did absolutely nothing to ensure data redundancy. May you suggest > me some practices that are considered as production ready?
Unfortunately there's no single up-to-date page that I can found that best answer the question for beginners. The best doc in ubuntu is probably https://wiki.ubuntu.com/Kernel/Reference/ZFS , which links to https://pthree.org/2012/12/13/zfs-administration-part-viii-zpool-best-practices-and-caveats/ (warning: some entries might be outdated). LXD's use of file-backed pool is perfectly described on the last "caveats" entry: "Don't commit production data to file VDEVs. Only use file VDEVs for testing scripts or learning the ins and outs of ZFS." I tend to agree with Sean on this one though: the detailed discussion for this is more appropriate on zfsonlinux mailing list, not here. I just wish that ubuntu's docs would properly warn the use of file-backed pool. -- Fajar _______________________________________________ lxc-users mailing list [email protected] http://lists.linuxcontainers.org/listinfo/lxc-users
